Files
advanced-methods-practise-s…/w9/DimensionalityReduction.ipynb
2025-07-02 08:44:09 +02:00

220 lines
110 KiB
Plaintext

{
"cells": [
{
"cell_type": "markdown",
"id": "d4b15792",
"metadata": {},
"source": [
"# Programming Exercise: Dimensionality Reduction PCA vs AE\n",
"In this exercise you will have to reduce the dimensionality of the Iris dataset to 2 using PCA and an AE.\n",
"The AutoEncoder consists of an encoder (4→8→2) and a decoder (2→8→4)."
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "c4539f14",
"metadata": {},
"outputs": [],
"source": [
"#You can use these packages, but you can also use other\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"from sklearn.datasets import load_iris\n",
"from sklearn.preprocessing import StandardScaler\n",
"from sklearn.decomposition import PCA\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.linear_model import LogisticRegression\n",
"from sklearn.metrics import mean_squared_error, accuracy_score\n",
"\n",
"import tensorflow as tf\n",
"from tensorflow.keras.layers import Input, Dense\n",
"from tensorflow.keras.models import Model\n",
"\n",
"plt.rcParams['figure.figsize'] = (6,4)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "f50f2ef0",
"metadata": {},
"outputs": [],
"source": [
"# Please enter your names\n",
"name = \"Fabian Langer, Yannik Bretschneider\""
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "ac003f40",
"metadata": {},
"outputs": [],
"source": [
"#load data\n",
"iris = load_iris()\n",
"X = iris.data \n",
"y = iris.target \n",
"scaler = StandardScaler()\n",
"X_scaled = scaler.fit_transform(X)"
]
},
{
"cell_type": "markdown",
"id": "f8407c9f",
"metadata": {},
"source": [
"# PCA\n",
"Apply PCA to reduce the dimension to 2."
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "b370a7f8",
"metadata": {},
"outputs": [],
"source": [
"# PCA\n",
"pca = PCA(n_components=2)\n",
"Z_pca = pca.fit_transform(X_scaled)"
]
},
{
"cell_type": "markdown",
"id": "13a3def8",
"metadata": {},
"source": [
"# AE\n",
" Create and train an AutoEncoder consisting of an encoder (4→8→2) and a decoder (2→8→4).\n",
" Train it for 200 epochs with a batch size of 16."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "5d1e2b11",
"metadata": {},
"outputs": [
{
"ename": "ValueError",
"evalue": "Cannot convert '4' to a shape.",
"output_type": "error",
"traceback": [
"\u001b[31m---------------------------------------------------------------------------\u001b[39m",
"\u001b[31mValueError\u001b[39m Traceback (most recent call last)",
"\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[28]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[32m----> \u001b[39m\u001b[32m1\u001b[39m input_l = \u001b[43mInput\u001b[49m\u001b[43m(\u001b[49m\u001b[43mshape\u001b[49m\u001b[43m=\u001b[49m\u001b[32;43m4\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[32m 3\u001b[39m \u001b[38;5;66;03m# encoder (4→8→2)\u001b[39;00m\n\u001b[32m 4\u001b[39m encoded = Dense(\u001b[32m8\u001b[39m, activation=\u001b[33m'\u001b[39m\u001b[33mrelu\u001b[39m\u001b[33m'\u001b[39m)(input_l)\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/repos/advanced-methods-practise-sose2025/.venv/lib/python3.12/site-packages/keras/src/layers/core/input_layer.py:209\u001b[39m, in \u001b[36mInput\u001b[39m\u001b[34m(shape, batch_size, dtype, sparse, ragged, batch_shape, name, tensor, optional)\u001b[39m\n\u001b[32m 144\u001b[39m \u001b[38;5;129m@keras_export\u001b[39m([\u001b[33m\"\u001b[39m\u001b[33mkeras.layers.Input\u001b[39m\u001b[33m\"\u001b[39m, \u001b[33m\"\u001b[39m\u001b[33mkeras.Input\u001b[39m\u001b[33m\"\u001b[39m])\n\u001b[32m 145\u001b[39m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34mInput\u001b[39m(\n\u001b[32m 146\u001b[39m shape=\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[32m (...)\u001b[39m\u001b[32m 154\u001b[39m optional=\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[32m 155\u001b[39m ):\n\u001b[32m 156\u001b[39m \u001b[38;5;250m \u001b[39m\u001b[33;03m\"\"\"Used to instantiate a Keras tensor.\u001b[39;00m\n\u001b[32m 157\u001b[39m \n\u001b[32m 158\u001b[39m \u001b[33;03m A Keras tensor is a symbolic tensor-like object, which we augment with\u001b[39;00m\n\u001b[32m (...)\u001b[39m\u001b[32m 207\u001b[39m \u001b[33;03m ```\u001b[39;00m\n\u001b[32m 208\u001b[39m \u001b[33;03m \"\"\"\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m209\u001b[39m layer = \u001b[43mInputLayer\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m 210\u001b[39m \u001b[43m \u001b[49m\u001b[43mshape\u001b[49m\u001b[43m=\u001b[49m\u001b[43mshape\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 211\u001b[39m \u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m=\u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 212\u001b[39m \u001b[43m \u001b[49m\u001b[43mdtype\u001b[49m\u001b[43m=\u001b[49m\u001b[43mdtype\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 213\u001b[39m \u001b[43m \u001b[49m\u001b[43msparse\u001b[49m\u001b[43m=\u001b[49m\u001b[43msparse\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 214\u001b[39m \u001b[43m \u001b[49m\u001b[43mragged\u001b[49m\u001b[43m=\u001b[49m\u001b[43mragged\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 215\u001b[39m \u001b[43m \u001b[49m\u001b[43mbatch_shape\u001b[49m\u001b[43m=\u001b[49m\u001b[43mbatch_shape\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 216\u001b[39m \u001b[43m \u001b[49m\u001b[43mname\u001b[49m\u001b[43m=\u001b[49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 217\u001b[39m \u001b[43m \u001b[49m\u001b[43minput_tensor\u001b[49m\u001b[43m=\u001b[49m\u001b[43mtensor\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 218\u001b[39m \u001b[43m \u001b[49m\u001b[43moptional\u001b[49m\u001b[43m=\u001b[49m\u001b[43moptional\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 219\u001b[39m \u001b[43m \u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m 220\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m layer.output\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/repos/advanced-methods-practise-sose2025/.venv/lib/python3.12/site-packages/keras/src/layers/core/input_layer.py:92\u001b[39m, in \u001b[36mInputLayer.__init__\u001b[39m\u001b[34m(self, shape, batch_size, dtype, sparse, ragged, batch_shape, input_tensor, optional, name, **kwargs)\u001b[39m\n\u001b[32m 89\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[33m\"\u001b[39m\u001b[33mYou must pass a `shape` argument.\u001b[39m\u001b[33m\"\u001b[39m)\n\u001b[32m 91\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m shape \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[32m---> \u001b[39m\u001b[32m92\u001b[39m shape = \u001b[43mbackend\u001b[49m\u001b[43m.\u001b[49m\u001b[43mstandardize_shape\u001b[49m\u001b[43m(\u001b[49m\u001b[43mshape\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m 93\u001b[39m batch_shape = (batch_size,) + shape\n\u001b[32m 95\u001b[39m \u001b[38;5;28mself\u001b[39m._batch_shape = backend.standardize_shape(batch_shape)\n",
"\u001b[36mFile \u001b[39m\u001b[32m~/repos/advanced-methods-practise-sose2025/.venv/lib/python3.12/site-packages/keras/src/backend/common/variables.py:587\u001b[39m, in \u001b[36mstandardize_shape\u001b[39m\u001b[34m(shape)\u001b[39m\n\u001b[32m 585\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[33m\"\u001b[39m\u001b[33mUndefined shapes are not supported.\u001b[39m\u001b[33m\"\u001b[39m)\n\u001b[32m 586\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mhasattr\u001b[39m(shape, \u001b[33m\"\u001b[39m\u001b[33m__iter__\u001b[39m\u001b[33m\"\u001b[39m):\n\u001b[32m--> \u001b[39m\u001b[32m587\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[33mf\u001b[39m\u001b[33m\"\u001b[39m\u001b[33mCannot convert \u001b[39m\u001b[33m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mshape\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m'\u001b[39m\u001b[33m to a shape.\u001b[39m\u001b[33m\"\u001b[39m)\n\u001b[32m 588\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m config.backend() == \u001b[33m\"\u001b[39m\u001b[33mtensorflow\u001b[39m\u001b[33m\"\u001b[39m:\n\u001b[32m 589\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(shape, tf.TensorShape):\n\u001b[32m 590\u001b[39m \u001b[38;5;66;03m# `tf.TensorShape` may contain `Dimension` objects.\u001b[39;00m\n\u001b[32m 591\u001b[39m \u001b[38;5;66;03m# We need to convert the items in it to either int or `None`\u001b[39;00m\n",
"\u001b[31mValueError\u001b[39m: Cannot convert '4' to a shape."
]
}
],
"source": [
"input_l = Input(shape=(4,))\n",
"\n",
"# encoder (4→8→2)\n",
"encoded = Dense(8, activation='relu')(input_l)\n",
"encoded = Dense(2, activation='relu')(encoded)\n",
"\n",
"# decoder (2→8→4)\n",
"decoded = Dense(8, activation='relu')(encoded)\n",
"decoded = Dense(4, activation='linear')(decoded)\n",
"\n",
"# autoencoder: this model maps an input to its reconstruction\n",
"autoencoder = Model(input_l, decoded)\n",
"\n",
"# Separate encoder model: this model maps an input to its encoded representation\n",
"encoder = Model(input_l, encoded)\n",
"\n",
"# Compile the autoencoder\n",
"autoencoder.compile(optimizer='adam', loss='mean_squared_error')\n",
"\n",
"# Train autoencoder\n",
"autoencoder.fit(X_scaled, X_scaled,\n",
" epochs=200,\n",
" batch_size=16,\n",
" shuffle=True,\n",
" verbose=0)\n",
"\n",
"# Extract reduced output for X_scaled\n",
"Z_ae = encoder.predict(X_scaled)"
]
},
{
"cell_type": "markdown",
"id": "a6c6e7a1",
"metadata": {},
"source": [
"# Visualization"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "f71870c1",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 600x400 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 600x400 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# AE\n",
"plt.figure()\n",
"plt.scatter(Z_ae[:,0], Z_ae[:,1], c=y, cmap='plasma', s=40)\n",
"plt.title('AE 2-D Embedding')\n",
"plt.colorbar(label='species')\n",
"plt.show()\n",
"\n",
"# PCA\n",
"plt.figure()\n",
"plt.scatter(Z_pca[:,0], Z_pca[:,1], c=y, cmap='cividis', s=40)\n",
"plt.title('PCA 2-D Embedding')\n",
"plt.colorbar(label='species')\n",
"plt.show()\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": ".venv",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.7"
}
},
"nbformat": 4,
"nbformat_minor": 5
}